Personalization through Implicit Scoring has been one of the core features of Umbraco Engage since day one, but here's a secret: score-based personalization sometimes gets it wrong, and when it does, it can feel random to marketers.
In today's blog we will be taking a look at how implicit scoring works, where it starts to crack in the real world, and how Umbraco Engage's latest feature solves this!
How implicit personalization works
Implicit personalization is all about building confidence over time. As visitors move through your site, every action they take tells you a little more about who they are. Engage lets you assign points to these actions, gradually mapping behavior to a persona or a step in a customer journey. Once a visitor crosses a certain score threshold, the system decides: "okay, this person fits persona X" or "they've reached step Y." And from that point, you can start personalizing their experience.
You've got a few different ways to build up that score. You can award points based on what content someone is looking at, which external site or page referred them, what campaign brought them in, or whether they triggered a specific goal. And if none of those fit your use case, you can wire up your own custom scoring, maybe pulling data from a CRM or some other external system. More information about Engage's scoring system can be found on the Umbraco Docs.
Where implicit scoring falls short
Sometimes there is a gap between what the system thinks and what the marketer knows. Imagine a visitor scores 51 arbitrary points for "Coffee Lover" and 49 for "Tea Enthusiast", is that really a confident classification? A single pageview can swing the active persona and influence the direction you are sending that visitor towards.
What about a returning visitor? Someone who scored heavily for "Coffee Lover" six months ago might have completely different interests now, but those old points are still sitting there, quietly influencing their experience.
Or perhaps the scenario where high-intent actions are drowned out by noise? A visitor fills out your enterprise contact form, which is about as clear a signal as you can get. But because they also casually browsed ten pages of unrelated content, the scoring puts them in a completely different persona. The marketer knows who this person is, but the system disagrees.
None of this means implicit scoring is broken. For most visitors, it works well. But it highlights that a confidence-based system is, by definition, probabilistic. And sometimes probabilistic isn't good enough.
That's where explicit scoring comes in
With the latest release of Umbraco Engage 17.1.0 we introduce a new feature called Explicit Scoring, or 'locking'. Instead of relying on building confidence over time, sometimes you know a visitor clicked a very specific link, completed a very specific goal, or was manually identified by the sales team. In those moments, probabilistic scoring is overkill. You want to say: "This person IS a Coffee Lover, period. Stop guessing"
We have added the ability to lock that confidence directly onto the same actions that would usually just add up points. 🔒
Attachment 1. Locking of Customer Journeys
When a visitor triggers an action with the intent of locking them to that Persona or Customer Journey Step, that Persona or Step will be applied to them immediately. This can then be used to personalize content right away with confidence that the visitor is classified correctly.
Some examples of this could be a specific mail campaign that only classifies you as a Coffee Lover, or perhaps an Umbraco Form that identifies you as an Enterprise customer straight away. That lock stays in place indefinitely. The only things that remove it are the visitor triggering another locked action in the same group, or a marketer manually toggling it off for that specific visitor.
Finding the balance
So should you lock everything? Definitely not. Implicit scoring exists for a good reason. Most of your visitors will never fill out that enterprise form or click that one specific link. For the majority of your traffic, gradually learning from behavior is exactly the right approach. It lets the system learn without requiring any manual intervention, and for most use cases, it gets it right.
But for that subset of visitors where you do have a clear signal, where an action leaves no room for interpretation, locking removes the guesswork entirely. Think of it less as replacing the scoring system and more as giving marketers a safety net. The scoring engine handles the 80%, and explicit locks handle the 20% where you already know the answer. ⚖️
Summary
Implicit scoring is a powerful tool that works well for the majority of your visitors, but it's not perfect. For those moments where a visitor's intent is crystal clear, Explicit Scoring gives you the ability to lock that classification in with confidence. Together they give you the best of both worlds: a system that learns on its own, with the option to step in when you know better.
If you've made it all the way to the bottom, I would like to thank you for your interest, and if there is anything I've mentioned that you'd like to learn more about feel free to contact me over at my socials available at the Contact page, of which I may expand on in a future blog! 😊
